Degrees of freedom

Degrees of Freedom (DF) can be defined as the number of observations in a sample, minus the number of parameters measured from the sample (Davies 1974). What this means in practical terms, is that every time a parameter (such as the mean of a sample of data) is used in a calculation of another value (such as the standard deviation) the number of degrees of freedom will decrease by 1.

The reason for this has to do with the independence of samples and the information they convey in the calculation.

In the statistic, the degrees of freedom can have a maximum value equal to the number of segments used  - 1.

is the number of parameters which must be measured from the sample data: 2 for a normal distribution and 3 for a log normal distribution. In addition, at least one degree of freedom is lost through the calculation of the difference between the observed and expected values. Additional degrees of freedom are lost when segments are amalgamated and other parameters are calculated.